Memory replacement

[Arnd Bergmann]

On Friday 11 March 2011, John Watlington wrote:
> On Mar 9, 2011, at 2:23 PM, Arnd Bergmann wrote:
> 
> > On Wednesday 09 March 2011 17:31:24 Kevin Gordon wrote:
> >> go, no-go, spend the extra pennies and get a Class 4/6/8/10
> > 
> > Note that Class 8 does not exist (except fakes) and class 10 is
> > usually not faster than class 6 if you run ext3 on it.
> > 
> > Also, a Sandisk card is usually faster than a card from
> > most other manufacturers even if they are one class faster
> > nominally.
> 
> I'll call BS on that claim.   Sandisk cards are all over the map,
> depending on the controller used internally.    Please understand
> that these manufacturers change controllers all the time --- tests
> results from nine months ago are invalid.

I've tested around a dozen media from them, and while you are true
that they use rather different algorithms and NAND chips inside, all
of them can write to at least 5 erase blocks before getting into
garbage collection, which is really needed for ext3 file systems.

Contrast this with Kingston cards, which all use the same algorithm
and can only write data linearly to one erase block at a time, resulting
in one or two orders of magnitude higher internal write amplification.

Most other vendors are somewhere inbetween, and you sometimes get
fake cards that don't do what you expect, such a a bunch of Samsung
microSDHC cards thaI have I which are labeled Sandisk on the outside.

I've also seen some really cheap noname cards outperform similar-spec'd
sandisk card, both regarding maximum throughput and the garbage collection
algorithms, but you can't rely on that.

	Arnd